Accurately estimating individualized probability of developing breast cancer over time is clinically useful for early detection and prevention. InterGenetics Incorporated has developed OncoVue(R), a model incorporating multiple, commonly occurring genetic variants and personal history information, to estimate age-specific risk of developing sporadic breast cancer. This model consisting of 22 SNPs in 19 genes and several epidemiological risk factors was developed utilizing multivariate logistic regression analysis on a geographically diverse large case-control dataset (training set) genotyped for a comprehensive panel of common single nucleotide polymorphisms (SNPs) known or highly likely to influence tumorigenesis. Using a number of measures of performance applied to both the training set and two independent test sets, OncoVue exhibited improved performance in identifying women that are truly at higher risk for breast cancer (previously diagnosed breast cancer cases) compared to the widely utilized Gail model. Although the current model has improved clinical utility, our long-term goal is to continue to refine OncoVue by identifying and incorporating additional novel genetic variants associated with breast cancer risk. Approximately 10% of the SNP polymorphisms examined in our original cancer-related candidate SNP panel were not included in building the OncoVue model because they were out of Hardy-Weinberg equilibrium in the controls. Recently, it has become clear that most of these candidate gene polymorphisms that proved technically challenging occur in copy number variable (CNV) regions of the genome. This widespread and newly discovered source of genetic variation encompasses regions of DNA that can range in size from thousands to millions of base pairs. Genome wide analyses show that ~12% of the human genome exhibits CNV and that ~10% of gene containing regions are subject to CNV. CNV can make a substantial contribution to differences between individuals with a significant number of genes present in three or more copies per genome, in contrast to the conventional two copies per genome. Furthermore, recently published studies on the candidate gene SULT1A1 clearly demonstrated that CNV underlies observed individual variation in enzymatic activity in vivo. Thus, CNVs represent an important and novel source of genetic variation with strong potential for association with predisposition to developing complex diseases such as breast cancer. The proposed research will investigate genomic CNV in candidate genes known or likely to have a role in breast cancer development with the long-term goal of utilizing newly discovered associations to improve the ability of OncoVue to estimate individualized risk. We propose the following specific aims for this Phase I project: (1) To complete a pilot study to determine the frequency of occurrence of CNV across 20 candidate genes. We will examine these candidate genes for CNV in DNAs from 120 individuals consisting of 40 Caucasian, 40 African-American and 40 Hispanic cancer-free women enrolled in our current studies. (2) To determine if CNV in selected candidates is associated with breast cancer risk. Candidate genes identified in Aim 1 with CNV at a frequency of e1% will be examined in breast cancer cases and cancer-free controls from our existing larger case-control study. Potential associations will be investigated in a multiethnic panel including Caucasian (500 Ca/500 Co), African American (250 Ca/250 Co) and Hispanic (100 Ca/ 100 Co) samples representing a randomly generated subset of our larger study. The successful completion of this Phase I study has the potential to discover novel associations of CNV with susceptibility to developing breast cancer. In Phase II, we will be poised to expand these studies to genotype our larger case-control collection and incorporate associations with these novel polymorphisms into the next generation of tests for risk of breast and other cancers. PUBLIC HEALTH RELEVANCE: InterGenetics Incorporated has developed OncoVue(R), a model incorporating multiple, commonly occurring genetic variants and personal history information, to estimate age-specific risk of developing sporadic breast cancer. Such a model has clinical utility in identifying patients that will benefit most from preventive measures and increased surveillance which may facilitate early detection and prevention. The proposed research will investigate the potential association of copy number variation (CNV), a newly recognized and widespread source of variation in the human genome, with predisposition to breast cancer in Caucasians, African Americans and Hispanics enrolled in our large case-control study. The long term goal of these studies is to improve and optimize the ability of OncoVue to accurately estimate individualized risk of developing breast cancer by identifying and integrating additional informative markers into the model. [unreadable] [unreadable] [unreadable] [unreadable]